Sihyun Jeon

Various Graph Neural Networks (GNNs) have been successful in analyzing data in non-Euclidean spaces, however, they have limitations such as oversmoothing, i.e., information becomes excessively averaged as the number of hidden layers increases. The issue stems from the intrinsic formulation of conventional graph convolution where the nodal features are aggregated from a direct neighborhood per layer across the entire nodes in the graph. As setting different number of hidden layers per node is infeasible, recent works leverage a diffusion kernel to redefine the graph structure and incorporate information from farther nodes. Unfortunately, such approaches suffer from heavy diagonalization of a graph Laplacian or learning a large transform matrix. In this regards, we propose a diffusion learning framework, where the range of feature aggregation is controlled by the scale of a diffusion kernel. For efficient computation, we derive closed-form derivatives of approximations of the graph convolution with respect to the scale, so that node-wise range can be adaptively learned. With a downstream classifier, the entire framework is made trainable in an end-to-end manner. Our model is tested on various standard datasets for node-wise classification for the state-of-the-art performance, and it is also validated on a real-world brain network data for graph classifications to demonstrate its practicality for Alzheimer classification.

Detecting marine objects inshore presents challenges owing to algorithmic intricacies and complexities in system deployment. We propose a difficulty-aware edge-cloud collaborative sensing system that splits the task into object localization and fine-grained classification. Objects are classified either at the edge or within the cloud, based on their estimated difficulty. The framework comprises a low-power device-tailored front-end model for object localization, classification, and difficulty estimation, along with a transformer-graph convolutional network-based back-end model for fine-grained classification. Our system demonstrates superior performance (mAP@0.5 +4.3%}) on widely used marine object detection datasets, significantly reducing both data transmission volume (by 95.43%) and energy consumption (by 72.7%}) at the system level. We validate the proposed system across various embedded system platforms and in real-world scenarios involving drone deployment.

Various Graph Neural Networks (GNN) have been successful in analyzing data in non-Euclidean spaces, however, they have limitations such as oversmoothing, i.e., information becomes excessively averaged as the number of hidden layers increases. The issue stems from the intrinsic formulation of conventional graph convolution where the nodal features are aggregated from a direct neighborhood per layer across the entire nodes in the graph. As setting different number of hidden layers per node is infeasible, recent works leverage a diffusion kernel to redefine the graph structure and incorporate information from farther nodes. Unfortunately, such approaches suffer from heavy diagonalization of a graph Laplacian or learning a large transform matrix. In this regards, we propose a diffusion learning framework where the range of feature aggregation is controlled by the scale of a diffusion kernel. For efficient computation, we derive closed-form derivatives of approximations of the graph convolution with respect to the scale, so that node-wise range can be adaptively learned.With a downstream classifier, the entire framework is made trainable in an end-to-end manner. Our model is tested on various standard datasets for node-wise classification for the state-of-the-art performance, and it is also validated on a real-world brain network data for graph classifications to demonstrate its practicality for Alzheimer classification.

This study was conducted to identify factors related to loneliness during the coronavirus disease-2019 (COVID-19) pandemic and focused on how the fear of COVID-19 infection affects loneliness and the conditional effect of gratitude as a moderator in the relationship among the fear of COVID-19, anxiety, and loneliness.For the analysis of this study, a survey was conducted among 1,500 individuals aged 19 to 69 years living in three metropolitan areas in South Korea. Questionnaires included sociodemographic data, psychological experience and stress associated with COVID-19, Generalized Anxiety Disorder Scale-7, UCLA Loneliness Scale-3, and Gratitude Questionnaire-6. An analysis was conducted by applying SPSS PROCESS macro models 4 and 7.First, anxiety mediated the relationship between the fear of COVID-19 infection and loneliness. Second, the effect of the fear of COVID-19 infection on loneliness through anxiety was moderated by gratitude. The higher the gratitude, the more the fear of COVID-19 infection is buffered in the path to anxiety, and the lower the indirect effect on loneliness.This suggests that in the context of the COVID-19 pandemic, interventions for psychological problems such as anxiety and loneliness can be carried out through gratitude, a significant protective variable.

Recent strides in deep learning have yielded impres-sive practical applications such as autonomous driving, natural language processing, and graph reasoning. However, the sus-ceptibility of deep learning models to subtle input variations, which stems from device imperfections and non-idealities, or adversarial attacks on edge devices, presents a critical challenge. These vulnerabilities hold dual significance-security concerns in critical applications and insights into human-machine sen-sory alignment. Efforts to enhance model robustness encounter resource constraints in the edge and the black box nature of neural networks, hindering their deployment on edge devices. This paper focuses on algorithmic adaptations inspired by the human brain to address these challenges. Hyper Dimensional Computing (HDC), rooted in neural principles, replicates brain functions while enabling efficient, noise-tolerant computation. HDC leverages high-dimensional vectors to encode information, seamlessly blending learning and memory functions. Its trans-parency empowers practitioners, enhancing both robustness and understanding of deployed models. In this paper, we introduce the first comprehensive study that compares the robustness of HDC to white-box malicious attacks to that of deep neural network (DNN) models and the first HDC gradient-based attack in the literature. We develop a framework that enables HDC models to generate gradient-based adversarial examples using state-of-the-art techniques applied to DNNs. Our evaluation shows that our HDC model provides, on average, 19.9% higher robustness than DNNs to adversarial samples and up to 90% robustness improvement against random noise on the weights of the model compared to the DNN.

Abstract. The Aerodyne Aerosol Mass Spectrometer (AMS) and Aerosol Chemical Speciation Monitor (ACSM) are the most widely applied tools for in situ chemical analysis of the non-refractory bulk composition of fine atmospheric particles. The mass spectra (MS) of many AMS and ACSM observations from field and laboratory studies have been reported in peer-reviewed literature and many of these MS have been submitted to an open-access website. With the increased reporting of such datasets, the database interface requires revisions to meet new demands and applications. One major limitation of the web-based database is the inability to automatically search the database and compare previous MS with the researcher's own data. In this study, a searchable database tool for the AMS and ACSM mass spectral dataset was built to improve the efficiency of data analysis using Igor Pro, consistent with existing AMS and ACSM software. The database tool incorporates the published MS and sample information uploaded on the website. This tool allows the comparison of a target mass spectrum with the reference MS in the database, calculating cosine similarity, and provides a range of MS comparison plots, reweighting, and mass spectrum filtering options. The aim of this work is to help AMS and ACSM users efficiently analyze their own data for possible source or atmospheric processing features by comparison to previous studies, enhancing information gained from past and current global research on atmospheric aerosol.

Brain-inspired Hyper-dimensional computing (HDC) has recently shown promise as a lightweight machine learning approach. Despite its success, there are limited studies on the robustness of HDC models to adversarial attacks. In this paper, we introduce the first comparative study of the robustness between HDC and deep neural network (DNN) to malicious attacks. We develop a framework that enables HDC models to generate gradient-based adversarial examples using state-of-the-art techniques applied to DNNs. Our evaluation shows that HDC with a proper neural encoding module provides significantly higher robustness to adversarial attacks than existing DNNs. In addition, HDC models have high robustness to adversarial samples generated for DNNs.

Abstract. The Aerodyne Aerosol Mass Spectrometer (AMS) and Aerosol Chemical Speciation Monitor (ACSM) are the most widely applied tools for in-situ chemical analysis of the non-refractory bulk composition of fine atmospheric particles. The mass spectra (MS) of many AMS and ACSM observations from field and laboratory studies have been reported in peer-reviewed literature and many of these MS have been submitted to an open-access website. With the increased reporting of such data sets, the database interface requires revisions to meet new demands and applications. One major limitation of the web-based database is the inability to automatically search the database and compare previous MS with the researcher’s own data. In this study, a searchable database tool for the AMS and ACSM mass spectral dataset was built to improve the efficiency of data analysis using Igor Pro, consistent with existing AMS and ACSM software. The database tool incorporates the published MS and sample information uploaded on the website. This tool allows the comparison of a target mass spectrum with the reference MS in the database, calculating match similarity, and provides a range of MS comparison plots, reweighting, and mass spectrum filtering options. The aim of this work is to help AMS users efficiently analyze their own data for possible source or atmospheric processing features by comparison to previous studies, enhancing information gained from past and current global research on atmospheric aerosol.

Shower development dynamics for a jet traveling through the quark-gluon plasma (QGP) is a multiscale process, where the heavy flavor mass is an important scale. During the high virtuality portion of the jet evolution in the QGP, emission of gluons from a heavy flavor is modified owing to heavy quark mass. Medium-induced radiation of heavy flavor is sensitive to microscopic processes (e.g. diffusion), whose virtuality dependence is phenomenologically explored in this study. In the lower virtuality part of shower evolution, i.e. when the mass is comparable to the virtuality of the parton, scattering and radiation processes of heavy quarks differ from light quarks. The effects of these mechanisms on shower development in heavy flavor tagged showers in the QGP is explored here. Furthermore, this multiscale study examines dynamical pair production of heavy flavor (via virtual gluon splittings) and their subsequent evolution in the QGP, which is not possible otherwise. A realistic event-by-event simulation is performed using the JETSCAPE framework. Energy-momentum exchange with the medium proceeds using a weak coupling recoil approach. Using leading hadron and open heavy flavor observables, differences in heavy versus light quark energy-loss mechanisms are explored, while the importance of heavy flavor pair production is highlighted along with future directions to study.

This study examines the implications of the work in which theoretical thinking is visualized, realized, and embodied through art, focusing on the case of cinematic collaboration between cultural researchers and artists. To this end, the film <Frantz Fanon: Black Skin, White Mask> by British black artist Isaac Julien is compared and analyzed with the theoretical thinking of cultural researcher Stuart Hall. Stuart Hall has interacted with black artists since the 1980s, especially in the cinematic collaboration of Isaac Julien, who has shared a sense of the problem of “visualization of theory,” in his research on representation, race, and identity. Stuart Hall's thinking, which emphasized the politics of identification that moves in the characteristics of difference, self-reflection, and context dependence, is reproduced in this film around sight and boundaries. Through this, it can be confirmed that theoretical concepts that change according to history and context find a place to talk in the area of representation.

<strong class="journal-contentHeaderColor">Abstract.</strong> The Aerodyne Aerosol Mass Spectrometer (AMS) and Aerosol Chemical Speciation Monitor (ACSM) are the most widely applied tools for in-situ chemical analysis of the non-refractory bulk composition of fine atmospheric particles. The mass spectra (MS) of many AMS and ACSM observations from field and laboratory studies have been reported in peer-reviewed literature and many of these MS have been submitted to an open-access website. With the increased reporting of such data sets, the database interface requires revisions to meet new demands and applications. One major limitation of the web-based database is the inability to automatically search the database and compare previous MS with the researcher&rsquo;s own data. In this study, a searchable database tool for the AMS and ACSM mass spectral dataset was built to improve the efficiency of data analysis using Igor Pro, consistent with existing AMS and ACSM software. The database tool incorporates the published MS and sample information uploaded on the website. This tool allows the comparison of a target mass spectrum with the reference MS in the database, calculating match similarity, and provides a range of MS comparison plots, reweighting, and mass spectrum filtering options. The aim of this work is to help AMS users efficiently analyze their own data for possible source or atmospheric processing features by comparison to previous studies, enhancing information gained from past and current global research on atmospheric aerosol.

<strong class="journal-contentHeaderColor">Abstract.</strong> The Aerodyne Aerosol Mass Spectrometer (AMS) and Aerosol Chemical Speciation Monitor (ACSM) are the most widely applied tools for in-situ chemical analysis of the non-refractory bulk composition of fine atmospheric particles. The mass spectra (MS) of many AMS and ACSM observations from field and laboratory studies have been reported in peer-reviewed literature and many of these MS have been submitted to an open-access website. With the increased reporting of such data sets, the database interface requires revisions to meet new demands and applications. One major limitation of the web-based database is the inability to automatically search the database and compare previous MS with the researcher&rsquo;s own data. In this study, a searchable database tool for the AMS and ACSM mass spectral dataset was built to improve the efficiency of data analysis using Igor Pro, consistent with existing AMS and ACSM software. The database tool incorporates the published MS and sample information uploaded on the website. This tool allows the comparison of a target mass spectrum with the reference MS in the database, calculating match similarity, and provides a range of MS comparison plots, reweighting, and mass spectrum filtering options. The aim of this work is to help AMS users efficiently analyze their own data for possible source or atmospheric processing features by comparison to previous studies, enhancing information gained from past and current global research on atmospheric aerosol.

<strong class="journal-contentHeaderColor">Abstract.</strong> The Aerodyne Aerosol Mass Spectrometer (AMS) and Aerosol Chemical Speciation Monitor (ACSM) are the most widely applied tools for in-situ chemical analysis of the non-refractory bulk composition of fine atmospheric particles. The mass spectra (MS) of many AMS and ACSM observations from field and laboratory studies have been reported in peer-reviewed literature and many of these MS have been submitted to an open-access website. With the increased reporting of such data sets, the database interface requires revisions to meet new demands and applications. One major limitation of the web-based database is the inability to automatically search the database and compare previous MS with the researcher&rsquo;s own data. In this study, a searchable database tool for the AMS and ACSM mass spectral dataset was built to improve the efficiency of data analysis using Igor Pro, consistent with existing AMS and ACSM software. The database tool incorporates the published MS and sample information uploaded on the website. This tool allows the comparison of a target mass spectrum with the reference MS in the database, calculating match similarity, and provides a range of MS comparison plots, reweighting, and mass spectrum filtering options. The aim of this work is to help AMS users efficiently analyze their own data for possible source or atmospheric processing features by comparison to previous studies, enhancing information gained from past and current global research on atmospheric aerosol.

Various pion and photon production mechanisms in high-energy nuclear collisions at RHIC and LHC are discussed. Comparison with RHIC data is done whenever possible. The prospect of using electromagnetic probes to characterize quark-gluon plasma formation is assessed.

We document the activities performed during the second MadAnalysis 5 workshop on LHC recasting, that was organised in KIAS (Seoul, Korea) on February 12-20, 2020. We detail the implementation of 12 new ATLAS and CMS searches in the MadAnalysis 5 Public Analysis Database, and the associated validation procedures. Those searches probe the production of extra gauge and scalar/pseudoscalar bosons, supersymmetry, seesaw models and deviations from the Standard Model in four-top production.

The nuclear modification factor R_AA for charged hadron production at the LHC is predicted from jet energy loss induced by gluon bremsstrahlung. The Arnold, Moore, and Yaffe formalism is used, together with an ideal hydrodynamical model.

We present the activities performed during the first MadAnalysis 5 workshop on LHC recasting that has been organized at High 1 (Gangwon privince, Korea) on August 20-27, 2017. This report includes details on the implementation in the MadAnalysis 5 framework of eight ATLAS and CMS analyses, as well as a description of the corresponding validation and the various issues that have been observed.

"Hybrid Hadronization" is a new Monte Carlo package to hadronize systems of partons. It smoothly combines quark recombination applicable when distances between partons in phase space are small, and string fragmentation appropriate for dilute parton systems, following the picture outlined by Han et al. [PRC 93, 045207 (2016)]. Hybrid Hadronization integrates with PYTHIA 8 and can be applied to a variety of systems from $e^++e^-$ to $A+A$ collisions. It takes systems of partons and their color flow information, for example from a Monte Carlo parton shower generator, as input. In addition, if for $A+A$ collisions a thermal background medium is provided, the package allows sampling thermal partons that contribute to hadronization. Hybrid Hadronization is available for use as a standalone code and is also part of JETSCAPE since the 2.0 release. In these proceedings we review the physics concepts underlying Hybrid Hadronization and demonstrate how users can use the code with various parton shower Monte Carlos. We present calculations of hadron chemistry and fragmentation functions in small and large systems when Hybrid Hadronization is combined with parton shower Monte Carlos MATTER and LBT. In particular, we discuss observable effects of the recombination of shower partons with thermal partons.